Classification of Impressionist and Pointillist Paintings Based on Their Brushstrokes Characteristics

Author:

Georgoulaki Kristina1ORCID

Affiliation:

1. Department of Informatics and Computer Engineering, University of West Attica, Athens, Greece

Abstract

The classification of works of art in terms of artistic style is a complex task. Some painting styles are closely related to the form of their brushstrokes. Salient examples are Pointillism and Impressionism, having both distinguishable brushstroke characteristics which are small, rounded of clear color, repetitive dots for Pointillism style and visible, elongated and slanting, repetitive touches for Impressionism style. As Impressionism is the ancestral style of Pointillism, the two styles have many elements in common and distinguishing them is difficult. In this article, specific texture features are investigated for the classification of the two styles, focusing mainly on small differences in their brushstrokes. The texture features adopted are Granulometric features, gray-level co-occurrence matrix features, and run length features. It is shown experimentally that the run length method outperforms the other features and can efficiently (up to 95%) discriminate the two textured styles since it incorporates information about size, direction, and intensity of brushstrokes.

Publisher

Association for Computing Machinery (ACM)

Reference24 articles.

1. T. Putri, R. Mukundan, and K. Neshatian. 2017. Artistic style characterization of Vincent Van Gogh’s paintings using extracted features from visible brush strokes. In International Conference on Pattern Recognition Applications and Methods.

2. A. Lecoutre, B. Négrevergne, and F. Yger. 2017. Recognizing art style automatically in painting with deep learning. In Ninth Asian Conference on Machine Learning.

3. Classification of Pointillist paintings using colour and texture features

4. Oguz Icoglu, B. Gunsel, and Sanem Sariel. 2004. Classification and indexing of paintings based on art movements. In 12th European Signal Processing Conference. 749–752.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3